Connect AI assistants to Groww's live API using this custom MCP server to run real-time mutual fund filters, manager profiling, risk metrics, and portfolio diagnostics.
Story Behind the Project:
A few weeks ago, I wanted to do some deep research on mutual funds to find top-performing schemes and compare risk parameters.
I tried using Claude to speed up the process, hoping it could analyze different funds and give me quick insights. But relying on standard web searches turned into a frustrating experience. The AI kept hopping from one URL to another, pulling fragmented or outdated data from random financial blogs, and taking forever just to gather basic metrics.
I kept thinking, there has to be a better way. Why can't the AI just pull structured, live mutual fund data directly from a single reliable source?
Since I already use Groww, looking for their mutual fund API felt like the logical next step. Unfortunately, Groww only provides public APIs for stock trading, not mutual fund data. I searched around for existing MCP tools, but none of them covered Groww mutual funds either.
So I decided to build one myself.
I opened up Developer Tools, inspected the client-side network traffic while browsing Groww, and mapped out their underlying backend endpoints. Using those live endpoints, I built a custom Model Context Protocol (MCP) server from scratch using FastMCP.
Now, my research process is completely transformed. Instead of spending hours browsing financial portals, I can just ask Claude to screen, compare, and analyze any mutual fund on Groww in natural language.
Because the MCP server feeds live Groww data directly into Claude, the AI can intelligently screen funds using complex filters, inspect top 10 company holdings, evaluate fund manager track records, check risk metrics like Sharpe, Sortino, Alpha, and Beta, and compare CAGR against benchmark indices in seconds.
About Project
This project provides an intelligent, professional-grade Model Context Protocol (MCP) server that connects AI assistants like Claude directly to Groww's live API. It enables real-time mutual fund filtering, manager profiling, risk metric analysis, and portfolio diagnostics.
When you ask your AI assistant for mutual fund recommendations or analysis, it can autonomously translate your request into filter criteria, query Groww's backend in real-time, and perform deep portfolio analysis to identify the best mutual funds for your needs.
Key Capabilities:Direct Live Groww API Data: Fetches real-time, production-grade financial data directly from Groww.Top 10 Portfolio Company Holdings: Extracts top invested companies, sector allocation, and corpus percentage.Fund Manager Profiling: Retrieves fund managers' names, educational background, and career experience.Advanced Volatility & Risk Metrics: Evaluates Sharpe Ratio, Sortino Ratio, Alpha, Beta, and Standard Deviation.CAGR & Benchmark Comparison: Compares returns against benchmarks and category averages.Clean LLM Optimization: Prunes raw API JSON responses to save token context while preserving essential analytical data.Filter Capabilities Available to AI:Groww Verdict: TOP_BUY, BUY, HOLD, SELLAsset Categories: Equity, Debt, Hybrid, CommoditiesSub-Categories: Includes various sub-categories for Equity, Debt, Hybrid, and Commodities.Risk Profile: Low, Moderately Low, Moderate, Moderately High, High, Very HighSorting Options: Popularity, Prime Verdict, 1Y Returns, 3Y Returns, 5Y ReturnsAMC / Fund Houses: Supports over 50 fund houses.Index Only: Option to filter for Index Funds only.Installation & Authentication:
Clone the repository and configure authentication credentials in auth.py by extracting them from your browser's developer tools after logging into Groww.
The server can be registered with Claude Desktop for seamless integration. Configuration details for Windows and macOS are provided.
MCP Tools Overview:search_mutual_funds: Screens live mutual fund schemes from Groww based on provided criteria.fetch_mutual_fund_details: Retrieves in-depth holdings, risk metrics, returns, and fund manager profiles using the fund's search_id.Disclaimer: This project is an open-source tool for personal research and analysis and is not affiliated with Groww.
License: MIT License.
Built with